Structural, magnetic and theoretical calculations of a ferromagnetically coupled tetranuclear copper(ii) square complex

Article abstract of DOI:10.1039/c3nj01512b

Condensation of 3-hydroxysalicylaldehyde with 2,4,6-trimethyl-1,3- phenylenediamine gives the ligand N,N′-bis(3-hydroxysalicylidene)-2,4,6- trimethyl-m-phenylenediamine (H₄L). The dinuclear zinc(ii) complex [Zn(H₂L)]₂ (1) and the tetranuclear copper(ii) square complex [{Cu(H₂L)}₄(THF)] (2) were synthesized and structurally characterized by single crystal X-ray diffraction. In 1, the Zn^II ions are bridged by two H₂L molecules and they adopt a tetrahedral geometry. The coordination geometry of the copper atoms in 2 is either square planar or square pyramidal, each metal atom being bound to the NO donor sets of two Schiff base ligands and, in one case, to an extra THF molecule. The arrangement of the tetranuclear complexes in the crystal lattice results in the formation of square channels. Variable temperature magnetic measurements on 2 evidence significant long-range ferromagnetic interactions between the four Cu^II centres leading to an S = 2 ground state with J₁ = +5.81, J₂ = +2.36, J₃ = +1.73 and J ₄ = +2.37 cm^-1. DFT calculations have been carried out in order to corroborate the experimental fitting and ascertain the origin of this ferromagnetic behaviour.

Full text of DOI:10.1039/c3nj01512b

NJC
View Article Online
View Journal | View Issue
PAPER
Structural, magnetic and theoretical calculations
of a ferromagnetically coupled tetranuclear
copper(II) square complex†
Cite this: New J. Chem., 2014,
38, 1306
Lionel Salmon,*a Pierre Thuery,b Eric Riviere,c Jean-Pierre Costes,a
´
`
Antonio J . Motad and Michel Ephritikhineb
Condensation of 3-hydroxysalicylaldehyde with 2,4,6-trimethyl-1,3-phenylenediamine gives the ligand
N,N0-bis(3-hydroxysalicylidene)-2,4,6-trimethyl-m-phenylenediamine (H4L). The dinuclear zinc(II) complex
[Zn(H2L)]2 (1) and the tetranuclear copper(II) square complex [{Cu(H2L)}4(THF)] (2) were synthesized and
structurally characterized by single crystal X-ray diffraction. In 1, the ZnII ions are bridged by two H2L
molecules and they adopt a tetrahedral geometry. The coordination geometry of the copper atoms in 2 is
either square planar or square pyramidal, each metal atom being bound to the NO donor sets of two
Schiff base ligands and, in one case, to an extra THF molecule. The arrangement of the tetranuclear
complexes in the crystal lattice results in the formation of square channels. Variable temperature magnetic
measurements on 2 evidence significant long-range ferromagnetic interactions between the four CuII
centres leading to an S = 2 ground state with J1 = +5.81, J2 = +2.36, J3 = +1.73 and J4 = +2.37 cmꢀ1. DFT
calculations have been carried out in order to corroborate the experimental fitting and ascertain the origin
of this ferromagnetic behaviour.
Received (in Montpellier, France)
2nd December 2013,
Accepted 17th January 2014
DOI: 10.1039/c3nj01512b
www.rsc.org/njc
the magnetic orbitals is weak. The origin of the interaction can
be explained by delocalisation and/or polarisation phenomena.
Introduction
The design and synthesis of polynuclear coordination compounds In the latter case, the McConnell spin polarization mechanism6
and in particular copper compounds have always interested describes how an unpaired electron on one atom polarizes the
inorganic chemists working in the field of magnetochemistry.1 electron on the adjacent atom through the bridging ligand. DFT
These compounds can act as models for the multimetal active calculations7 are often used to explain and predict the nature
sites of metalloproteins2 and as precursors for high-nuclearity and the strength of these different behaviours.
clusters for the preparation of useful magnetic materials.3 The
In the course of our studies of magnetic interactions in
predominant factor in such systems is the control of the heteropolymetallic 3d–5f and homopolynuclear 5f Schiff base
magnetic exchange interaction J between magnetic centres; the sign complexes,8,9 we prepared the new bicompartmental hexadentate
and the amplitude of J are sensitive to small structural variations in ligand N,N0-bis(3-hydroxysalicylidene)-2,4,6-trimethyl-m-phenylene-
the bridge linking the metal centres. The most common interaction diamine, H4L (Scheme 1), with the aim of obtaining bis-dinuclear
through extended bridges is antiferromagnetic, because of the [M(3d)–U]2 compounds. Our step-by-step approach consists of
overlapping of the magnetic orbitals.4 However, ferromagnetic
coupling can be observed when the magnetic orbitals show exten-
sive spatial overlap, but are orthogonal,5 or when the overlapping of
a Laboratoire de Chimie de Coordination, CNRS UPR 8241, 205 Route de Narbonne,
31077 Toulouse Cedex 4, France. E-mail: lionel.salmon@lcc-toulouse.fr;
Fax: +33 05 61 55 30 03; Tel: +33 05 61 33 31 78
b CEA, IRAMIS, SIS2M, CNRS UMR 3299, CEA/Saclay, 91191 Gif-sur-Yvette, France
c
´
´
Laboratoire de Chimie Inorganique, Institut de Chimie Moleculaire et des Materiaux,
´
CNRS UMR 8613, Universite de Paris-Sud, 91405 Orsay Cedex, France
d
´
´
Departamento de Quımica Inorganica, Facultad de Ciencias,
Universidad de Granada, Avda. Fuentenueva s/n, 18002-Granada, Spain
† CCDC 909567 and 909568. For crystallographic data in CIF or other electronic
format see DOI: 10.1039/c3nj01512b
Scheme 1 The Schiff base H4L under study.
1306 | New J. Chem., 2014, 38, 1306--1314
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
Paper
NJC
were measured on a MPMS5 magnetometer (Quantum Design).
The powdered and dried samples of the compounds were
introduced in capsules before the introduction in the magneto-
meter. The calibration was made at 298 K by using a palladium
reference supplied by Quantum Design. The independence of
the susceptibility value with regard to the applied field was
checked at room temperature. The wMT data were collected over
the 2–300 K range at magnetic fields of 1 and 10 kOe and were
corrected for diamagnetism. Elemental analyses were per-
formed by Analytische Laboratorien at Lindlar (Germany).
Synthesis
Fig. 1 View of the complex [Zn(H2L)]2 (1). The carbon-bound hydrogen
atoms are omitted. The displacement ellipsoids are drawn at the 50%
The Schiff base ligand N,N0-bis(3-hydroxysalicylidene)-2,4,6-
trimethyl-m-phenylenediamine (H4L) was synthesized by con-
densation of 2 equivalents of 3-hydroxy-salicylaldehyde with
2,4,6-trimethyl-m-phenylenediamine in methanol. 1H NMR
(THF-d8, 23 1C): d = 1.98 (s, 3H, CH3), 2.12 (s, 6H, CH3), 6.73
(t, 2H, J = 7.8 Hz, aromatic H), 6.85 (d, 2H, J = 8.0 Hz, aromatic
H), 6.89 (d, 2H, J = 8.0 Hz, aromatic H), 7.00 (s, 1H, aromatic H),
7.97 (s, 2H, OH), 8.37 (s, 2H, CHQN) and 13.06 (s, 2H, OH). The
acac compounds Cu(acac)2 and Zn(acac)2 (Aldrich) were used
without purification.
[Zn(H2L)]2ꢁ3THF (1ꢁ3THF). H4L (200 mg, 0.51 mmol) was
mixed in a 1 : 1 ratio with Zn(acac)2 (144 mg, 0.51 mmol) in
THF, which led to the formation of a yellow powder at room
temperature. Recrystallization at 80 1C afforded yellow crystals
of [Zn(H2L)]2ꢁ3THF (85 mg, 37% yield). 1H NMR (THF-d8, 23 1C):
d = 1.97 (s, 3H, CH3), 2.09 (s, 6H, CH3), 6.53 (d, 2H, J = 8.2 Hz,
aromatic H), 6.59 (t, 2H, J = 8.0 Hz, aromatic H), 6.86 (d, 2H,
J = 8.1 Hz, aromatic H), 6.95 (s, 1H, aromatic H), 7.95 (s, 2H, OH) and
8.06 (s, 2H, CHQN). The elemental analyses are in agreement with
the unsolvated [Zn(H2L)]2 complex. Anal. calcd for C46H40N4O8Zn2:
C, 60.8; H, 4.4; N, 6.2. Found: C, 58.9; H, 4.8; N, 6.0.
0
probability level. Symmetry code: = 1 – x, 1 – y, 1 – z.
synthesizing first the mononuclear M(3d) complex and then the
heterodinuclear M(3d)–U compound. The reaction of Zn(acac)2
with H4L led to the formation of the expected bis-mononuclear
complex [Zn(H2L)]2ꢁ3THF (1ꢁ3THF); see Fig. 1. On the other
hand, the same reaction with the corresponding copper salt did
not afford the bis-copper complex but led to the formation of a
tetranuclear square complex [{Cu(H2L)}4(THF)]ꢁ4THF (2ꢁ4THF);
see Fig. 2. Investigation of the magnetic properties of 2ꢁ4THF
revealed the presence of long range intramolecular ferromagnetic
interactions between copper atoms. Several examples of ferro-
magnetically coupled tetranuclear copper(II) complexes have
been reported so far, most often with a Cu4O4 cubane-like
structure, or built from two Cu2 subunits.3,4,10 Moreover, ferro-
magnetic interaction was also reported for dinuclear copper(II)
entities chelated by Schiff base ligands similar to the one used
in this work.11
Experimental
General information
The 1H NMR spectra were recorded on a Bruker DPX 200 instrument
and referenced internally using the residual protio solvent reso-
nances relative to tetramethylsilane (d 0). Magnetic susceptibilities
[{Cu(H2L)}4(THF)]ꢁ4THF (2ꢁ4THF). H4L (300 mg, 0.77 mmol)
was mixed in a 1 : 1 ratio with Cu(acac)2 (201 mg, 0.77 mmol) in
THF, which led to the formation of a dark brown powder at
room temperature. Recrystallization at 80 1C afforded dark
brown crystals of [{Cu(H2L)(THF)}{Cu(H2L)}3]ꢁ4THF (220 mg,
57% yield). The elemental analyses are in agreement with
the [{Cu(H2L)(THF)}{Cu(H2L)}3]ꢁ2THF complex; Anal. calcd for
C
104H104N8O19Cu4: C, 61.7; H, 5.1; N, 5.5. Found: C, 61.5; H, 5.0;
N, 5.4.
Crystallography
The data were collected at 100(2) K on a Nonius Kappa-CCD
area detector diffractometer12 using graphite-monochromated
Mo Ka radiation (l = 0.71073 Å). The crystals were introduced in
glass capillaries with a protecting ‘‘Paratone-N’’ oil (Hampton
Research) coating. The unit cell parameters were determined
from 10 frames, and they were then refined on all data. The
data (j scans with 21 steps) were processed using HKL2000.13
Absorption effects were corrected empirically with the program
DELABS in PLATON.14 The structures were solved by direct
methods using SHELXS-97 and subsequent Fourier-difference
synthesis and refined by full-matrix least-squares on F2 using
Fig. 2 View of the complex [{Cu(H2L)}4(THF)]ꢁ4THF (2ꢁ4THF). The solvent
molecules, methyl substituents and carbon-bound hydrogen atoms are
omitted. The displacement ellipsoids are drawn at the 10% probability level. SHELXL-97.15 All non-hydrogen atoms were refined with anisotropic
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem., 2014, 38, 1306--1314 | 1307

View Article Online
NJC
Paper
Table 1 Crystal data and structure refinement details
1ꢁ3THF
dinuclear compound [Zn(H2L)]2ꢁ3THF (1ꢁ3THF) in 37% yield,
according to eqn (1).
2ꢁ4THF
THF
2H4L þ 2ZnðacacÞ2
½ZnðH2LÞꢂ2 þ4acacH
(1)
Empirical formula
C58H64N4O11Zn2 C112H120Cu4N8O21
ƒ!
Mr/g molꢀ1
1123.88
Monoclinic
P21/n
2168.32
Monoclinic
P21/n
ð1Þ
Crystal system
Space group
The centrosymmetric compound 1ꢁ3THF, with half a complex
molecule and 1.5 THF solvent molecule in the asymmetric unit,
is shown in Fig. 1, and selected bond lengths and angles are
given in Table 2. The two zinc atoms are doubly bridged by the
two diimino-m-phenylene groups of each ligand and the dis-
tance between the metal centres is 7.183(3) Å. The coordination
geometry around the zinc atoms can be described as distorted
tetrahedral, with angles around the metal atom spanning the
range 96.6(3)–127.7(3)1 and with a dihedral angle between the
two ZnNO planes of the ZnN2O2 core of 79.6(2)1. Such a
geometry is similar to that encountered in the corresponding
dinuclear cobalt complex [Co2(sal-m-phen)2] (H2sal-m-phen =
N,N0-bis(salicylidene)-m-phenylenediamine)23 with a dihedral
angle of 84.2(1)1 but it differs from those in the dinuclear
copper complexes [Cu2(sal-m-phen)2]24 and [Cu2(L0)2] (H2L0 =
N,N0-bis(2-hydroxy-3-methoxybenzylidene)-m-phenylenediamine),25
with dihedral angles of 43.6 and 51.5(1)1, respectively, for which
the geometry is intermediate between cis-planar and tetrahedral.
The mean values of the Zn–O and Zn–N bond lengths of 1.921(9)
and 2.016(2) Å, respectively, are slightly longer, by ca. 0.3 Å,
a/Å
b/Å
c/Å
b/1
V/Å3
Z
12.235(3)
16.9476(18)
13.042(3)
103.294(9)
2631.8(9)
2
1.418
0.977
1176
16 859
13.2695(8)
26.7073(11)
29.3020(17)
99.835(3)
10231.8(10)
4
1.408
0.896
4528
60 892
18 131
11 073
0.113
1338
0.081
r
calcd/g cmꢀ3
m(Mo-Ka)/mmꢀ1
F(000)
Reflections collected
Independent reflections
Observed reflections [I > 2s(I)] 2167
4594
Rint
0.116
355
0.094
0.220
1.010
ꢀ0.50
0.76
Parameters refined
R1
wR2
0.225
1.021
S
Drmin/e Åꢀ3
Drmax/e Åꢀ3
ꢀ0.70
1.08
displacement parameters. The hydrogen atoms bound to
oxygen atoms were found on a Fourier-difference map and all
the others were introduced at calculated positions (except for
the disordered THF molecule in 1ꢁ3THF). All were treated as
riding atoms with an isotropic displacement parameter equal
to 1.2 times that of the parent atom (1.5 for CH3). In 1ꢁ3THF,
one THF molecule is disordered over two symmetry-related
positions sharing two carbon atoms, which have been refined
with occupancy parameters of 0.5 and some restraints on
displacement parameters. In 2ꢁ4THF, one carbon atom in
two THF molecules is disordered over two positions which
have been refined with occupancy parameters constrained to
sum to unity and restraints on bond lengths; restraints on
displacement parameters were applied for the atoms of the
THF molecules.
Table 2 Selected bond lengths (Å) and angles (1) in compounds 1ꢁ3THF
and 2ꢁ4THF
1ꢁ3THF
Zn–O2
Zn–O30
Zn–N1
Zn–N20
1.930(6)
1.912(6)
2.018(8)
2.014(7)
O2–Zn–N1
N1–Zn–N20
N20–Zn–O30
O30–Zn–O2
O2–Zn–N20
N1–Zn–O30
96.7(3)
111.1(3)
96.6(3)
108.2(3)
127.7(3)
118.3(3)
Znꢁ ꢁ ꢁZn0
7.183(3)
0
Symmetry code: = 1 – x, 1 – y, 1 – z
2ꢁ4THF
Cu1–O2A
Cu1–O3D
Cu1–N1A
Cu1–N2D
Cu1–O1
1.904(5)
1.905(5)
2.002(6)
1.991(6)
2.405(6)
7.1706(12)
N1A–Cu1–O2A
O2A–Cu1–N2D
N2D–Cu1–O3D
O3D–Cu1–N1A
N1A–Cu1–N2D
O2A–Cu1–O3D
91.6(2)
90.0(2)
90.6(2)
90.6(2)
152.5(2)
174.1(2)
Crystal data and structure refinement details are given in
Table 1. The molecular plots were drawn using ORTEP-316 and
VESTA.17
Cu1ꢁ ꢁ ꢁCu2
Computational details. All theoretical calculations were carried
out at the density functional theory (DFT) level using the hybrid
B3LYP exchange–correlation functional,18 as implemented in the
Gaussian 09 program.19 A quadratic convergence method was
employed in the self-consistent-field process.20 The triple-z quality
basis set proposed by Ahlrichs et al. has been used for all atoms.21
Calculations were performed on complexes built from experi-
mental geometries as well as on model complexes. The electronic
configurations used as starting points were created using the
Jaguar 7.9 software.22
Cu2–O2B
Cu2–O3A
Cu2–N1B
Cu2–N2A
1.881(5)
1.892(4)
2.008(6)
2.003(6)
N1B–Cu2–O2B
O2B–Cu2–N2A
N2A–Cu2–O3A
O3A–Cu2–N1B
N1B–Cu2–N2A
O2B–Cu2–O3A
92.1(2)
91.5(2)
92.0(2)
90.1(2)
154.3(2)
167.1(2)
Cu2ꢁ ꢁ ꢁCu3
6.8691(11)
Cu3–O2C
Cu3–O3B
Cu3–N1C
Cu3–N2B
1.894(5)
1.903(5)
1.999(6)
1.996(6)
N1C–Cu3–O2C
O2C–Cu3–N2B
N2B–Cu3–O3B
O3B–Cu3–N1C
N1C–Cu3–N2B
O2C–Cu3–O3B
91.8(2)
89.3(2)
91.4(2)
89.8(2)
159.9(2)
173.5(2)
Cu3ꢁ ꢁ ꢁCu4
6.9683(12)
Cu4–O2D
Cu4–O3C
Cu4–N1D
Cu4–N2C
1.891(5)
1.896(5)
1.997(6)
1.987(6)
N1D–Cu4–O2D
O2D–Cu4–N2C
N2C–Cu4–O3C
O3C–Cu4–N1D
N1D–Cu4–N2C
O2D–Cu4–O3C
92.7(2)
89.6(2)
92.6(2)
90.8(2)
161.3(2)
162.5(2)
Results and discussion
Syntheses and characterization
Treatment of H4L with 1 mol equivalent of Zn(acac)2 (acac =
MeCOCHCOMe) in refluxing THF afforded yellow crystals of the
Cu4ꢁ ꢁ ꢁCu1
6.8876(12)
1308 | New J. Chem., 2014, 38, 1306--1314
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
Paper
NJC
Table 3 Hydrogen bonding geometry (Å, 1) in compounds 1ꢁ3THF and
2ꢁ4THF
D–Hꢁ ꢁ ꢁA
D–H
Hꢁ ꢁ ꢁA
Dꢁ ꢁ ꢁA
D–Hꢁ ꢁ ꢁA
1ꢁ3THF
O1–H1ꢁ ꢁ ꢁO2
O1–H1ꢁ ꢁ ꢁO20
O4–H4ꢁ ꢁ ꢁO3
O4–H4ꢁ ꢁ ꢁO6
0.94
0.94
0.98
0.98
2.13
2.12
2.26
2.08
2.680(8)
2.751(9)
2.658(9)
2.81(2)
116
123
103
130
0
Symmetry code: = 1 – x, 1 – y, 1 – z
2ꢁ4THF
O1A–H1Aꢁ ꢁ ꢁO2A
O4A–H4Aꢁ ꢁ ꢁO3A
O1B–H1Bꢁ ꢁ ꢁO2B
O4B–H4Bꢁ ꢁ ꢁO3B
O1C–H1Cꢁ ꢁ ꢁO2C
O4C–H4Cꢁ ꢁ ꢁO3C
O1D–H1Dꢁ ꢁ ꢁO2D
O4D–H4Dꢁ ꢁ ꢁO3D
0.86
0.91
0.90
0.88
0.91
0.92
0.98
0.91
2.19
2.09
2.07
2.12
2.08
2.10
2.07
1.99
2.624(6)
2.625(7)
2.624(7)
2.631(7)
2.641(7)
2.660(7)
2.627(7)
2.620(7)
111
116
119
116
119
118
114
125
Fig. 3 View of the complex [{Cu(H2L)}4(THF)]ꢁ4THF (2ꢁ4THF). The carbon-
bound hydrogen atoms and two THF molecules are omitted. The displace-
ment ellipsoids are drawn at the 10% probability level.
than in the corresponding copper complexes,24,25 in agreement
with the increase in the ionic radius when passing from copper to
zinc with a coordination number of 4.26 Both planar salicylidene-
iminato fragments form dihedral angles of 80.2(3)1 with the
bridging phenylene rings to which they are attached. The two
central phenylene rings in the molecule correspond to an inter-
planar separation of 3.35 Å, but the large offset between the
centroids (3.08 Å) precludes the presence of a p-stacking inter-
action. Atoms O1 and O4 are both protonated, which results in
the formation of intra- and intermolecular bifurcated hydrogen
bonds with the neighbouring phenoxide and THF solvent oxygen
atoms (Table 3). The intermolecular hydrogen bonds between
atoms O1 and O2 from neighbouring molecules result in the
formation of chains directed along the c axis.
Contrary to the dinuclear zinc compound 1 and the pre-
viously reported copper24,25 and cobalt23 congeners, dark brown
crystals of the tetranuclear compound [{Cu(H2L)}4(THF)]ꢁ4THF
(2ꢁ4THF) were isolated in 57% yield from the reaction of H4L
with 1 mol equivalent of Cu(acac)2 in refluxing THF, according
to eqn (2). Such a difference is difficult to explain, resulting
either from the ionic radii of the 3d metal atoms, the substitu-
tion at the meta position of the salicylaldimine rings, or the
reaction conditions (reflux in the present case).
being present in the crystal since the space group is centro-
symmetric. Three of the four copper atoms have a distorted
square planar geometry, being bound to two salicylaldiminato
moieties from two adjacent Schiff base ligands and the fourth
copper atom has a square pyramidal geometry, with the addi-
tional oxygen atom of a THF molecule, O1, in the apical
position. The dihedral angles between the two halves of the
CuN2O2 core are 28.1(3), 28.5(2), 21.0(2) and 25.1(2)1 for
Cu1–Cu4, respectively, the copper atoms being localized at
0.189(3), 0.118(3), 0.121(3) and 0.018(3) Å from the mean
N2O2 plane (rms deviations 0.29, 0.33, 0.23 and 0.31 Å). The
mean values of the Cu–O(Schiff base) and Cu–N bond lengths of
1.896(8) and 1.998(6) Å, respectively, are in the range of those
measured in the dinuclear copper counterparts.24,25 The cyclic
architecture of the whole assembly can be seen as an open cage
with the bis-salicylideneiminato/copper fragments defining the
four sides and the central phenylene rings of the ligands defining
the vertices, with a methyl substituent pointing inwards. The
dihedral angles between the two arms of the ligands, which form
two adjacent faces of the cage, are not far from right angles [mean
value 81(3)1]. The bridging phenylene rings form pairs of dihedral
angles with the adjacent CuNO planes with mean values of 75.8,
77.7, 72.0 and 82.31 when going from Cu1 to Cu4, respectively.
The mean value of the intramolecular Cuꢁ ꢁ ꢁCu distances between
adjacent cations is 6.97(12) Å. The volume of the inner cavity can
THF
4H4L þ 4CuðacacÞ2
½fCuðH2LÞg4ðTHFÞꢂ þ8acacH (2)
ƒ!
ð2Þ
Complex 2, represented in Fig. 2 and 3, is a tetranuclear square be estimated from the distances between two opposite copper
complex devoid of any true symmetry element. However, if the atoms, 10.0433(13) and 9.6471(12) Å, and the distance between
coordinated THF molecule is disregarded, a pseudo four-fold the more distant carbon atoms of the two salicylaldimino
rotation axis bisecting the lines joining the opposite copper fragments in the third direction, ca. 10.9 Å. The resulting cavity
atoms is present, together with four orthogonal pseudo two-fold size, about 1000 Å3, accommodates two THF solvent molecules
rotation axes, two of them containing opposite copper atoms (Fig. 2). Moreover, as in the dinuclear zinc compound, atoms
and the two others containing the centroids of opposite O1 and O4 of the four ligands are protonated, which results
bridging aromatic rings. The complex thus presents an approxi- in intramolecular hydrogen bonds with the neighbouring
mate D4 point symmetry group, with the four ligands arranged phenoxide oxygen atoms (Table 3). The shortest intermolecular
in a helical fashion around the copper atom square; the Cuꢁ ꢁ ꢁCu separations, Cu3ꢁ ꢁ ꢁCu10 and Cu2ꢁ ꢁ ꢁCu400, of 6.5435(12)
molecule is consequently chiral, with both right and left forms and 6.1202(12) Å, respectively, are associated with the presence
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem., 2014, 38, 1306--1314 | 1309

View Article Online
NJC
Paper
Fig. 5 Thermal dependence of wMT for complex [{Cu(H2L)}4(THF)] (2). The
solid line was generated from the best fit parameters given in the text. The
inset represents the coupling exchange pathways.
from the wMT = f (T) curve are J1 = +5.81, J2 = +2.36, J3 = +1.73,
J4 = +2.37 cmꢀ1, g = 2.02 and TIP = 246 ꢃ 10ꢀ6 cm3 molꢀ1 (see
Fig. 5). The fit considering only one J value was not satisfactory.
The magnetization data for 2, at T = 2 K, are shown in Fig. 6
as an M versus H plot, M being the molar magnetization per
four copper ions and H the applied magnetic field. As expected,
the isothermal curve is well matched by the Brillouin function
for an isolated S = 2 spin state with g = 2.02.
The origin of the ferromagnetism in 2 could not be attributed
to copper–copper intermolecular interaction because of the
long Cuꢁ ꢁ ꢁO intermolecular distances. Besides, as the unpaired
electron of the copper(II) ion is more delocalised towards the
four equatorial bonds, it is clear that the spin density on the
Fig. 4 Crystal packing in [{Cu(H2L)}4(THF)]ꢁ4THF (2ꢁ4THF) viewed down
the a axis. Hydrogen atoms are omitted.
of slight intermolecular contacts between neighbouring
molecules, Cu3ꢁ ꢁ ꢁO1A0 2.569(5) and Cu2ꢁ ꢁ ꢁO1D00 2.800(5) Å,
these oxygen atoms being in axial positions (symmetry codes:
0
00
= x ꢀ 0.5, 1.5 ꢀ y, z ꢀ 0.5; = 0.5 ꢀ x, 0.5 + y, 1.5 ꢀ z). Each
complex is thus connected to four of its neighbours to generate
%
a two-dimensional assembly parallel to (101). Within the packing,
the cyclic complexes are aligned so as to form channels parallel
to the a axis, which are occupied by THF molecules (Fig. 4).
No intermolecular hydrogen bonding occurs, all the hydrogens
of the free hydroxyl groups being bound to the neighbouring
coordinated oxygen atom on each catechol moiety in the same
molecule.
Magnetic properties
The temperature dependence of the wMT product for 2 in the
2–300 K temperature range is shown in Fig. 5. At room
temperature, wMT is equal to 1.65 cm3 molꢀ1 K, a value close to
that expected for four magnetically isolated spins (1.5 cm3 molꢀ1
K with g = 2). Upon cooling, the value of wMT increases, attaining
3 cm3 molꢀ1 K at 2 K. This behaviour is typical of a ferro-
magnetically coupled complex; the observed wMT maximum
value is close to the ordinary one for a ground state of
S = 2 (3.06 cm3 molꢀ1 K with g = 2.02).
The appropriate Hamiltonian for an isolated cyclic tetra-
nuclear compound can be written as in eqn (3), in which
S1 = S2 = S3 = S4 = 1/2.
ˆ
H = ꢀJ1S1S2 ꢀ J2S2S3 ꢀ J3S3S4 ꢀ J4S4S1
(3)
The magnetic susceptibility has been computed by using
Fig. 6 Field dependence of the magnetization for complex [{Cu(H2L)}4(THF)]
(2) at 2, 4 and 6 K. The solid lines correspond to the Brillouin functions for
´
the procedure developed by Borras-Almenar and co-workers
(MAGPACK program).27 The best fitting parameters obtained the S = 2 ground state.
1310 | New J. Chem., 2014, 38, 1306--1314
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
Paper
NJC
axial oxygen atom of the adjacent molecule should be very J3 = +7.2, J4 = +4.2, and diagonals, J5 = ꢀ0.008, J6 = +0.003, which
small. Consequently, the ferromagnetic coupling measured in are in agreement with the experimental values found for this
2 is operating in an intramolecular way through the meta- complex. From these values, it is clear that the diagonal con-
phenylenediamine bridge.
tribution is almost negligible and could be omitted from the
magnetic coupling pathway scheme, the main contribution
being the coupling along the square sides, which gives a small
ferromagnetic behaviour. Furthermore, these lateral J values are
all very similar, independent of the 4 or 4 + 1 copper coordina-
tion. Thus, the axially coordinated neutral ligand has a limited
influence on the basal magnetic orbital, as expected. That all
these Cu–Cu magnetic interactions are ferromagnetic in nature
have been ascertained by taking just one side of the square (that
corresponding to the J1 pathway, see Fig. 7), leading to a
dinuclear model structure (2intra), see Fig. 8 (left), in which all
the remaining terminal valences coming from the broken bonds
have been replaced by hydrogen atoms and the copper coordina-
tion set to 4 in both centres. The remaining geometrical para-
meters have been kept as they appear in the crystal structure.
From this building block we have then calculated a Jintra value of
+6.5 cmꢀ1, in agreement with the original value calculated for 2
(+5.5 cmꢀ1). Thus, the magnetic coupling between the copper
centres is ferromagnetic in nature and there is a very small effect
of the axially coordinated ligands on the J magnitude.
DFT calculations
In an effort to get complementary information on compound 2,
and in order to confirm the magneto-structural correlation men-
tioned above, we have performed DFT calculations following the
Broken-Symmetry (BS) approach from the model crystal structure
depicted in Fig. 7, in which the four different chemical environ-
ments of the copper centres have been considered. This local
asymmetry, therefore, leads to six different J values: four corres-
ponding to the square sides and two to the diagonals. These
values could, in turn, be calculated from eight different BS spin
states, namely, one quintuplet (+ + + +), four triplets (ꢀ + + +,
+ ꢀ + +, + + ꢀ +, and + + + ꢀ), and three singlets (ꢀ ꢀ + +, + ꢀ ꢀ +,
and ꢀ + ꢀ +). From the equations issued from these spin states
we got the following J values, in cmꢀ1: sides, J1 = +5.5, J2 = +6.3,
In addition, in order to evaluate the intermolecular magnetic
exchange, we have also calculated an approximate magnetic
coupling constant ( Jinter) between two Cu4 units by considering
just two copper centres at the square vertices, leading to a
model (2inter) consisting of a copper dimer (Fig. 8, right).
In this case, the calculated magnetic exchange constant is only
+0.04 cmꢀ1, and so, should have a very small impact on the
global ferromagnetic behaviour, as expected. Therefore, from a
magnetic point of view, we can consider the magnetic contribu-
tion as coming from isolated squares formed by tetracoordinated
copper(II) cations.
The spin density map constructed for the tetranuclear unit
(Fig. 9, left) indicates a possible source of the ferromagnetic
behaviour: a spin polarization mechanism28 operating between
the copper centres through the aromatic ligands (Fig. 9, right).
This mechanism is based on an alternation of the spin density
sign from the bonded ligand atoms, which suffer a direct exchange
mechanism, through the p-cloud of aromatic bridges (Table 4).
As expected, the main spin density lies on the copper ions,
Fig. 7 The tetranuclear structure (2) considered for the DFT calculations
and the corresponding J scheme.
Fig. 8 Dinuclear models established for calculating the nature of the magnetic coupling for intra- (left, 2intra), and intermolecular (right, 2inter
)
magnetic interactions.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem., 2014, 38, 1306--1314 | 1311

View Article Online
NJC
Paper
Fig. 9 Left: calculated spin densities for the tetranuclear model of complex 2 in the ground (quintuplet) state. The represented isodensity surface
corresponds to a cut-off value of 0.0008 e bohrꢀ3. Grey and blue colors correspond to positive and negative values, respectively. Right: scheme of the
polarization mechanism operating between the copper centres showing the spin density relative values.
Table 4 Spin density values (in eꢀ) on selected atoms for the calculated
model of complex 2
Atomsa
Ground (quintuplet) state
Cu1/Cu2/Cu3/Cu4
N1/N2/N3/N4b
O1/O2/O3/O4b
OTHF
+0.6240/+0.5848/+0.6118/+0.5829
+0.0841/+0.0952/+0.0884/+0.0963
+0.0936/+0.0944/+0.0922/+0.0951
+0.0010
Ohydroxyphenol
+0.0003
Fig. 10 Dinuclear model (2model) established for calculating the effect of
the Cu–N–C–C torsion angle value on the magnitude of the magnetic
coupling.
N–C–C–C–Nc
+0.0823/ꢀ0.0074/+0.0150/ꢀ0.0027/+0.0925
a
Indices 1, 2, 3 and 4 refer to the Cu1, Cu2, Cu3 and Cu4 atoms,
b
respectively. Mean values between each pair of N (imino) and O
c
(phenoxo) atoms bonded to each copper ion. Alternating spin density
values along the N–C–C–C–N pathway connecting the Cu1 and Cu2
centres.
magnetic coupling between the metallic centres is clearly ferro-
magnetic, and that the mechanism of the magnetic exchange
pathway corresponds well to a spin polarization mechanism.
These conclusions corroborate those reported for dinuclear copper
compounds with the same m-phenylene bridge.11a,b Moreover,
as proposed by G. Aromi and coworkers,11b the strength of the
magnetic coupling seems to be related to the electronic pro-
perties of the ligand and in particular to the sum of the
absolute values of the atomic spin population on the atoms
of the m–N–F–N moiety.
as they are the magnetic centres. Moreover, the spin density on
the metal-bonded oxygen atoms in the THF and o-hydroxyphenol
ligands completing the 4 + 1 copper coordination sphere is
nearly zero, thus corroborating the weak influence that these
axially positioned ligands could exert on the main basal mag-
netic pathways. Some representative spin density values are
collected in Table 4.
In the original crystal structure, the copper atoms are
displaced from the neighbouring trimethylbenzene planes, as
indicated by the values of the corresponding Cu–N–C–C torsion
angles, which differ by about 5 to 201 from the ideal value
Conclusion
of 901. In the case of a spin polarization mechanism, the value The Schiff base ligand N,N0-bis(3-hydroxysalicylidene)-2,4,6-
of this angle would be somewhat irrelevant. In order to corro- trimethyl-m-phenylenediamine (H4L) is efficient in stabilizing
borate this point, we have additionally prepared a new model either the dinuclear zinc(II) complex [Zn(H2L)]2 (1) or the tetra-
compound (2model) consisting also of a dinuclear compound nuclear copper(II) complex [{Cu(H2L)}4(THF)] (2). In the latter
derived from an edge of the square (as 2intra), but with idealized compound, the single crystal X-ray diffraction study shows an
ligands, angles and distances as shown in Fig. 10. From this unusual cyclic geometry with each copper atom bonded to the
model, we have symmetrically varied the torsion angle from 901 NO donor sets of two Schiff base ligands, the arrangement of
to 601 by 151 steps for both metallic moieties, obtaining a set of the neighbouring tetranuclear units in the crystal lattice resulting
calculated J parameters with very similar values despite the in the formation of square channels. The magnetic study on 2
high angular variation introduced: +2.9, +2.7 and +1.5 cmꢀ1
,
revealed ferromagnetic interactions between the four CuII centres
respectively. These results point out that the nature of the leading to an S = 2 ground state with J1 = +5.81, J2 = +2.36,
1312 | New J. Chem., 2014, 38, 1306--1314
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
Paper
NJC
J3 = +1.73 and J4 = +2.37 cmꢀ1. DFT calculations have been
carried out, which corroborate the experimental results and
ascertain the origin of this ferromagnetic behaviour. Moreover,
detailed analysis of the DFT calculation points out that the
mechanism of the magnetic exchange pathway is in agreement
with a spin polarization mechanism. Eventually, modifications
of this ligand could permit the increase of the strength of the
coupling exchange and/or the dimensionality of these materials
through the formation of bridges between the isolated tetra-
nuclear species so as to afford new metal–organic frameworks.
P. C. Junk, C. M. Kepert, F. Mabbs, B. Moubaraki,
K. S. Murray and L. Spiccia, Inorg. Chem., 2001, 40, 1536;
´
(d) M. Murugesu, R. Clerac, B. Pilawa, A. Mandel,
C. E. Anson and A. K. Powell, Inorg. Chim. Acta, 2002,
337, 328; (e) Y. Song, C. Massera, O. Roubeau, P. Gamez,
A. M. Manotti Lanfredi and J. Reedjik, Inorg. Chem., 2004,
43, 6842; ( f ) M. Fondo, A. M. Garcia-Deibe, M. Corbella,
E. Ruiz, J. Tercero, J. Sanmartin and M. R. Bermejo, Inorg.
Chem., 2005, 44, 5011; (g) E. A. Buvaylo, V. N. Kokozay,
O. Y. Vassilyeva, B. W. Skelton, J. Jezierska, L. C. Brunel and
A. Ozarowski, Inorg. Chem., 2005, 44, 206; (h) C. H. Weng,
S. C. Cheng, H. M. Wei, H. H. Wei and C. J. Lee, Inorg. Chim.
Acta, 2006, 359, 2029; (i) A. Burkhardt, E. T. Spielberg,
References
¨
H. Gorls and W. Plass, Inorg. Chem., 2008, 47, 2485;
( j) S. Thakurta, P. Roy, R. J. Butcher, M. Salah El Fallah,
J. Tercero, E. Garribba and S. Mitra, Eur. J. Inorg. Chem.,
2009, 4385; (k) M. Fondo, N. Ocampo, A. M. Garcia-Deibe
and J. Sanmartin, Eur. J. Inorg. Chem., 2010, 2376;
(l) X. Zhang, B. Li, J. Tang, J. Tian, G. Huang and
J. Zhang, Dalton Trans., 2013, 42, 3308; (m) Z. Lu,
T. Fan, W. Guo, J. Lu and C. Fan, Inorg. Chim. Acta, 2013,
400, 191.
1 (a) L. K. Thompson, Coord. Chem. Rev., 2002, 193, 233;
(b) O. Kahn, Molecular Magnetism, VCH, New York, 1993.
2 (a) E. I. Solomon, R. K. Szilagyi, S. Debeer George and
L. Basumallick, Chem. Rev., 2004, 104, 419; (b) G. Henkel
and B. Krebs, Chem. Rev., 2004, 104, 801.
3 (a) V. A. Milway, S. M. Tareque Abedin, L. K. Thompson and
¨ ¨
D. O. Miller, Inorg. Chim. Acta, 2006, 359, 2700; (b) P. Seppala,
¨¨
E. Colacio, A. J. Mota and R. Sillanpaa, Inorg. Chem., 2013,
11 (a) I. Fernandez, R. Ruiz, J. Faus, M. Julve, F. Lloret, J. Cano,
X. Ottenwaelder, Y. Journaux and M. Carmen Munoz,
Angew. Chem., Int. Ed., 2001, 40, 3039; (b) A. Ranjan Paital,
T. Mitra, D. Ray, W. Tak Wong, J. Ribas-Arino, J. J. Novoa,
J. Ribas and G. Aromi, Chem. Commun., 2005, 5172;
52, 11096 and reference therein.
4 (a) X. Zhang, B. Li, J. Tang, J. Tian, G. Huang and J. Zhang,
Dalton Trans., 2013, 42, 3308 and reference therein;
(b) E. F. Hasty, L. J. Wilson and D. N. Hendeickson,
Inorg. Chem., 1978, 17, 1834.
˜
(c) R. H. Laye and E. C. Sanudo, Inorg. Chim. Acta, 2009,
5 O. Kahn, J. Galy, Y. Journaux, J. Jaud and I. Morgenstern-
Badarau, J. Am. Chem. Soc., 1982, 104, 2165.
6 H. M. McConnell, A. Robert and A. Welch, Proc. Robert A.
Welch Found. Conf. Chem. Res., 1967, 11, 144.
362, 2205; (d) R. Pandey, P. Kumar, A. K. Singh, M. Shalid,
P. Z. Li, S. K. Singh, Q. Xu, A. Misra and D. S. Pandey,
Inorg. Chem., 2011, 50, 3189.
12 R. W. W. Hooft, COLLECT, Nonius BV, Delft, The Netherlands,
1998.
13 Z. Otwinowski and W. Minor, Methods Enzymol., 1997,
276, 307.
14 A. L. Spek, J. Appl. Crystallogr., 2003, 36, 7.
15 G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystallogr.,
2008, 64, 112.
7 (a) S. P. Foxon, G. M. Torres, O. Walter, J. Z. Pedersen,
H. Toftlund, M. Hu¨ber, K. Falk, W. Haase, J. Cano, F. Lloret,
M. Julve and S. Schindler, Eur. J. Inorg. Chem., 2004, 335;
(b) E. Ruiz, Principles and Applications of Density in Inorganic
Chemistry II Book Series: Structure and Bonding, 2004,
vol. 113, p. 71.
´
`
8 (a) L. Salmon, P. Thuery, E. Riviere, J. J. Girerd and
16 L. J. Farrugia, J. Appl. Crystallogr., 1997, 30, 565.
17 K. Momma and F. Izumi, J. Appl. Crystallogr., 2008,
41, 653.
M. Ephritikhine, Chem. Commun., 2003, 762; (b) L. Salmon,
´
`
P. Thuery, E. Riviere, J. J. Girerd and M. Ephritikhine, Dalton
´
`
Trans., 2003, 2872; (c) L. Salmon, P. Thuery, E. Riviere and
18 (a) A. D. Becke, Phys. Rev. A, 1988, 38, 3098; (b) C. T. Lee,
W. T. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter
Mater. Phys., 1988, 37, 785; (c) A. D. Becke, J. Chem. Phys.,
1993, 98, 5648.
M. Ephritikhine, Inorg. Chem., 2006, 45, 83; (d) L. Salmon,
´
P. Thuery and M. Ephritikhine, Polyhedron, 2007, 26, 631;
´
(e) L. Salmon, P. Thuery and M. Ephritikhine, Polyhedron,
2007, 26, 645.
´
9 (a) L. Salmon, P. Thuery and M. Ephritikhine, Polyhedron, 19 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
´
2004, 23, 623; (b) L. Salmon, P. Thuery and M. Ephritikhine,
M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr,
T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam,
S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi,
G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji,
M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai,
M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross,
C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski,
P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador,
´
Dalton Trans., 2004, 1635; (c) L. Salmon, P. Thuery and
M. Ephritikhine, Dalton Trans., 2004, 4139; (d) L. Salmon,
P. Thuery and M. Ephritikhine, Polyhedron, 2006, 1537;
(e) L. Salmon, P. Thuery, S. Miyamoto, T. Yamato and
´
´
M. Ephritikhine, New J. Chem., 2006, 30, 1220.
10 (a) X. S. Tan, Y. Fujii, R. Nukada, M. Mikuriya and
Y. Nakano, J. Chem. Soc., Dalton Trans., 1999, 2415;
¨
(b) E. Colacio, M. Ghazi, R. Kivekas and J. M. R. Moreno,
Inorg. Chem., 2000, 39, 2882; (c) B. Graham, M. T. W. Hearn,
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem., 2014, 38, 1306--1314 | 1313

View Article Online
NJC
Paper
´
´
J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. 23 R. Hernandez-Molina, A. Mederos, P. Gili, S. Domınguez,
´
Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D.
Rabuck, R. Raghavachari, J. B. Foresman, J. V. Ortiz,
F. Lloret, J. Cano, M. Julve, C. Ruiz-Perez and X. Solans,
J. Chem. Soc., Dalton Trans., 1997, 4327.
Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B. B. 24 C. A. Bear, J. M. Waters and T. N. Waters, J. Chem. Soc. A,
Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, 1970, 2494.
R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. 25 C. T. Zeyrek, A. Elmali, Y. Elerman and I. Svoboda,
Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, Z. Naturforsch., 2005, 60b, 143.
B. Johnson, W. Chen, M. W. Wong, C. Gonzalez and 26 R. D. Shannon, Acta Crystallogr., Sect. A: Fundam. Crystallogr.,
J. A. Pople, Gaussian 03, Revision C.02, Gaussian, Inc.,
Wallingford, CT, 2004.
1976, 32, 751.
´
27 J. J. Borras-Almenar, J. M. Clemente-Juan, E. Coronado and
20 G. B. Bacskay, Chem. Phys., 1981, 61, 385.
B. S. Tsukerblat, J. Comput. Chem., 2001, 22, 985.
¨
´
´
21 A. Schafer, C. Huber and R. Ahlrichs, J. Chem. Phys., 1994, 28 M. A. Palacios, A. Rodrıguez-Dieguez, A. Sironi, J. M. Herrera,
100, 5829.
22 Jaguar, version 7.9, Schrodinger, LLC, New York, NY, 2012.
A. J. Mota, J. Cano and E. Colacio, Dalton Trans., 2009, 8538
and references therein.
¨
1314 | New J. Chem., 2014, 38, 1306--1314
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014